We propose to develop and validate a technology that will enable unequivocal genome-wide functional identification of cancer-related genes using either a positive or negative selection in conditions of either low or high stringency. This will extend the reach of forward genetics to a principally new set of phenomena. Identification of the genes and their products, which form the molecular basis of cellular phenotypes, is the fundamental problem in molecular analysis of cancer. The cellular factors that control proliferation, death, response to therapy, motility, interaction with other cells and the environment have emerged as the diagnostic and prognostic markers, as well as the targets of therapeutic intervention. Forward genetics is a popular methodology that uses genetic tools to uncover the modulators of various biological processes. Insertional mutagenesis based on random insertion of a strong and regulated promoter is a versatile, cost-efficient, unbiased and comprehensive approach to forward genetics in somatic cells. However, the current implementations of this approach apply exclusively to the conditions of stringent positive selection, and are inapplicable to a wider range of clinically-significant phenomena. Moreover, large-scale mapping and validation of multiple relevant insertional targets remains the technical bottleneck of the whole approach. To overcome these limitations, we will develop a new technology, which combines the elements of insertional mutagenesis with those of microarray analysis. The microarray component will allow high-throughput mapping of the inserts in a highly complex pool of cells. Positive or negative changes in the representation of individual inserts could be used to identify the loci, which affect the cell behavior under various selective conditions. The dependence of the changes on the function of the inserted promoter could be used for large-scale validation of the findings. We will construct a series of appropriate lentiviral vectors and will confirm their activity as insertional mutagens. We will optimize the procedure to monitor the frequency of individual mutant alleles in a high-throughput format. Finally, we will apply the newly developed tools and procedures to the discovery of genes whose products determine the response of prostate carcinoma to Taxol, the only chemotherapeutic compound known to extend the life of patients with androgen-independent prostate cancer. We will identify the events that either sensitize or protect cells from the drug. The clinical significance of these findings will be pursued in future studies, while our data, tools and procedures will be shared with the research community.